User researchers from across government met yesterday to share stories and experiences about user research they’ve done in improving the tools we use within government to deliver services.
Researchers from the Home Office, Ministry of Justice, Land Registry and the Government Digital Service shared case studies. We heard about tools used by civil servants to check applicant eligibility, process applications, update circumstances, buy digital services in government, and tools used by civil servants throughout their career.
Despite the diversity of the case studies, some very similar themes emerged. In this post, I’ll share some of those themes.
Internal user research brings its own set of challenges
Several speakers talked about the challenges that recruiting internal user research participants can bring.
Recruiting is slower and more time consuming
It’s easy to assume that recruiting for internal research will be easy and fast because, “I’m surrounded by my participants, right?”
In general, it was agreed that recruiting for internal research is slower and more time consuming than recruiting for external user research. That is, unless you’re studying just one team and can behave more like an anthropologist.
GDS user researcher, Angela Collins-Rees:
With a recruitment agency we can specify when and where we want to meet/hold our sessions and for how long. We set the parameters and the agency works to find participants who can fit in with that. With internal users, it's a bit more of a bargaining game. We can't be as rigid. We're also not incentivising internal participants. We're relying on their goodwill and interest to give up their time and participate in our research.
In general, with internal research you have to do all the recruitment yourself and it demands constant effort and time. Potential participants often find it difficult to make time in their work day for you to visit, to get permission from their managers to participate, or to come to the user research lab.
It’s important to use a consent form
There are also sensitive, ethical issues to consider when doing internal user research.
It’s very important that you use a consent form so that participants are clear about how the information they share will and won’t be used. For example, participants often need assurance that the information shared will be used to inform the design of the system they’re using, but won’t be passed on to their line manager for performance management purposes.
Participants are aware that the information they’re sharing could impact their career. It’s important to be sensitive to this. Protect your internal research participants in that same way as you would your external participants.
Start to end. Front and back.
As user researchers, we need to make sure we don’t just look at the experiences users have of a service at a single point in the transaction. Follow the service from the very beginning - often before it reaches government, perhaps starting with professional advisors who have user needs we could do a better job of satisfying.
And it's not only about end to end, but also front of house to back of house.
Ministry of Justice user researcher, Ana Santos:
When delivering a service, it's important to consider the link between the internal and the publics' user needs of the service, so you're not creating a silo'ed service with mismatched expectations on either side. We spent a lot of time gathering needs from our internal users (court staff) and analysing how this impacted our public users. It's important to match both groups needs in the end to end service so that you're continuously working towards eliminating 'waste'.
Make sure to include not only end users (front of house) but also the people inside government delivering the service (back of house) so that you understand the real opportunities to improve the service - often the best opportunities for improvement are those that are invisible to the user.
Fixing internal tools and processes might be one of the best ways to improve the service for end users
User researchers regularly find that the greatest gains in improving a service can be made by better supporting the people inside government who are doing the hard work to deliver those services. By improving internal tools and processes, we can reduce the time needed to deliver the service to end users. This should relieve backlogs and, in turn, reduce waiting time for end users.
GDS user researcher, Flora Bowden:
People can feel left in the dark and unsure about what's happening while their application is being processed. It can be really concerning and people will want to chase up the progress or find out if something's gone wrong. If we can improve government's internal processes, we can speed up the waiting time and improve the service for end users.
By applying the same approach we use in improving end-user facing services to internal tools, we not only improve the working experience and efficiency of civil servants but also the experience our end users have of the service.
Doing user research for internal tools is a great way to identify opportunities to do more with less and improve the user experience all at once.
Related reading
Internet Hurting Productivity by Gerry McGovern
Technology Led Organisational Transformation – Powerful Agent for Change by Tom Read
Join the Cross Government User Research Community
The Cross Government User Research community meets regularly to share experiences and improve our capability for user research in government. If you do user research in government and would like to find out more, join our Cross Government User Research Community.
Keep in touch. Sign up to email updates from this blog. Follow Leisa on Twitter
1 comment
Comment by Ady Garrett posted on
I have a 4 year old cheap-ish netbook that struggles to perform at the best of times; if I had to rely on that to conduct my online affairs then I’d find it a very frustrating experience.
When designing GOV.UK online services, particularly those that require some sort of user interaction, e.g. filing a tax return, do you take such things into account to ensure that interaction uses the least computer resource possible?